Loading…
Loading…
Models / Qwen3-Coder-Next
2 models (~79.7B to ~79.7B) · ↓ 1.8M downloads a month · from $0.78/h on 4x RTX 4060 Ti right now. Each model's versions do the same job; quantized ones (AWQ, FP8, 4-bit) need less GPU memory and run on cheaper machines.
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| OriginalQwen/Qwen3-Coder-Next-FP8 | vLLM | 80 GB | 92.4 GB | $1.10/h 2x RTX 4090 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| OriginalQwen/Qwen3-Coder-Next | vLLM | 159 GB | 180.2 GB | $2.36/h 4x Nvidia A40 |
| FP8 dynamic (RedHatAI)RedHatAI/Qwen3-Coder-Next-FP8-dynamic | vLLM | 82 GB | 93.9 GB | $1.38/h 8x RTX 3090 |
| NVFP4 GB10 (saricles)saricles/Qwen3-Coder-Next-NVFP4-GB10Licence to accept | vLLM | 46 GB | ~57.2 GB | $0.89/h 2x Nvidia A40 |
| FP8 (unsloth)unsloth/Qwen3-Coder-Next-FP8 | vLLM | 80 GB | 92.4 GB | $1.10/h 2x RTX 4090 |
| NVFP4 (RedHatAI)RedHatAI/Qwen3-Coder-Next-NVFP4 | vLLM | 48 GB | 56 GB | $0.78/h 4x RTX 4060 Ti |
| AWQ (qtum)qtum/Qwen3-Coder-Next-AWQ | vLLM | 42 GB | 49.9 GB | $0.78/h 4x RTX 4060 Ti |
Model details come from Hugging Face and are refreshed daily; prices follow the live offers. GPU memory is GridPup's estimate: the weights, room for an 8,192-token conversation (chat models) and the runtime's own memory. Large models are split across the GPUs of one machine ("4x RTX 4090").